ignite-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sergey Chugunov (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (IGNITE-4209) Semaphore gets broken when node finishes normally and leaves topology
Date Thu, 10 Nov 2016 12:44:58 GMT

     [ https://issues.apache.org/jira/browse/IGNITE-4209?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Sergey Chugunov updated IGNITE-4209:
------------------------------------
    Description: 
h4. Notes
Bug is reproducible with latest version of code available in master branch.

As [~vladisav] suggested (see link to mailing list below) root cause of the bug may be that
current implementation handles *EVT_NODE_LEFT* and *EVT_NODE_FAILED* events in the same way
and puts semaphore to a broken state even when one of cluster nodes finishes normally and
leaves topology.

h4. Steps to reproduce
# Using *ExampleNodeStartup* class start two or more server nodes.
# Using *IgniteSemaphoreExample* class run example and let it finish (it should finish normally
and then terminate).
# Run *IgniteSemaphoreExample* once again.

h4. Expected behavior
Second execution of the same example finishes successfully.
h4. Actual behavior
Second execution hangs forever, node running example never finishes.

Developers mailing list discussion: [semaphore unexpected behavior|http://apache-ignite-developers.2346864.n4.nabble.com/IGNITE-4155-IgniteSemaphoreExample-unexpected-behavior-td12077.html].



  was:
h4. Notes
Bug is reproducible with latest version of code available in master branch.

As [~vladisav] suggested (see link to mailing list below) root cause of the bug may be that
current implementation handles *EVT_NODE_LEFT* and *EVT_NODE_FAILED* events in the same way
and puts semaphore to a broken state.

h4. Steps to reproduce
# Using *ExampleNodeStartup* class start two or more server nodes.
# Using *IgniteSemaphoreExample* class run example and let it finish (it should finish normally
and then terminate).
# Run *IgniteSemaphoreExample* once again.

h4. Expected behavior
Second execution of the same example finishes successfully.
h4. Actual behavior
Second execution hangs forever, node running example never finishes.

Developers mailing list discussion: [semaphore unexpected behavior|http://apache-ignite-developers.2346864.n4.nabble.com/IGNITE-4155-IgniteSemaphoreExample-unexpected-behavior-td12077.html].




> Semaphore gets broken when node finishes normally and leaves topology
> ---------------------------------------------------------------------
>
>                 Key: IGNITE-4209
>                 URL: https://issues.apache.org/jira/browse/IGNITE-4209
>             Project: Ignite
>          Issue Type: Bug
>            Reporter: Sergey Chugunov
>
> h4. Notes
> Bug is reproducible with latest version of code available in master branch.
> As [~vladisav] suggested (see link to mailing list below) root cause of the bug may be
that current implementation handles *EVT_NODE_LEFT* and *EVT_NODE_FAILED* events in the same
way and puts semaphore to a broken state even when one of cluster nodes finishes normally
and leaves topology.
> h4. Steps to reproduce
> # Using *ExampleNodeStartup* class start two or more server nodes.
> # Using *IgniteSemaphoreExample* class run example and let it finish (it should finish
normally and then terminate).
> # Run *IgniteSemaphoreExample* once again.
> h4. Expected behavior
> Second execution of the same example finishes successfully.
> h4. Actual behavior
> Second execution hangs forever, node running example never finishes.
> Developers mailing list discussion: [semaphore unexpected behavior|http://apache-ignite-developers.2346864.n4.nabble.com/IGNITE-4155-IgniteSemaphoreExample-unexpected-behavior-td12077.html].



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message